SDXL 1.0 VRAM requirements

01The files, and how much VRAM each needs

“Needed” = file + 1.2 GB working memory + 0.8 GB system reserve. The 6.94 GB checkpoint also holds the two text encoders and the VAE. During sampling only the UNet (about 5.1 GB at 16-bit, 2.6B parameters) has to sit in VRAM, so that is the figure used for the 16-bit file.

02By amount of VRAM

03Best GPU for SDXL

The cheapest cards (by launch price) that run it well, and every card sorted by memory: best GPU for SDXL → Planning bigger images or longer clips? Open the calculator →

04By graphics card

On RTX 40/50 GPUs the FP8 file is preferred over Q8_0 when both fit (hardware FP8). All verdicts are calculated; see how the numbers work.

05Text encoder, VAE and other files

The text encoder is inside the checkpoint, so there is nothing extra to download.

06Where the files go in ComfyUI

Standard ComfyUI folders. After copying files, press R in ComfyUI (or restart it) to refresh the lists. Some uploads need their uploader's own loader node — see the notes above.

07AMD, Intel and NVIDIA: which file types are fast

Every file type loads on every listed GPU, so the memory verdicts apply to all of them. What differs is speed: FP8 maths needs RTX 40/50 or RX 9000 (with ROCm 6.4+ and PyTorch 2.7+); ComfyUI's INT8 maths runs on NVIDIA and AMD, not on Intel; GGUF is unpacked on the fly on any GPU, which costs some speed. NVFP4 files are fast only on RTX 50. AMD runs ComfyUI on Windows through ROCm, Intel through PyTorch XPU; some custom nodes are NVIDIA-only. Source: ComfyUI model_management.py. AMD and Intel guide →

08Measured and reported results

Reported results are other people's numbers, copied as published, with a link. Settings, drivers and ComfyUI versions differ, so compare them with care. Send yours.

09Training a LoRA for SDXL

reported Figures as stated by each trainer's own documentation or official example configs, read 2026-09-25. “Stated minimum” = the docs call it a minimum; “example run” = a config or measured run at that size. They differ a lot because of settings: an 8-bit or 4-bit base model, block swapping and lower resolution all cut memory. All models →

10Every file tracked for SDXL

Single checkpoint file loaded with Load Checkpoint. params_b is total incl. text encoders (UNet alone ~2.6B).